Research On Balanced Topology With Scale Free Theory For Wireless Sensor Networks

Wireless sensor networks(WSNs)consist of hundreds of low-cost and battery-powered sensors.Due to the limited battery power,sensors die over time,which degrades the network density and impacts the reliability of network.This requires an optimized design of topologies for WSNs,so as to obtain better robustness and longer lifetime.Barab(?)si and Albert proposed the first scale-free model,i.e.the BA model,which was proven to be robust against random removal or failure of nodes.Recent advances in the studies of complex networks motivated the research that adopts scale-free topologies for WSNs.However,a scale-free network typically results in a small amount of nodes with much higher degree than the others,which causes unbalanced energy consumption and transmission pressure of the sensor nodes.Therefore,constructing an balanced topology of WSNs became the study focus recently to prolong the lifetime of networks.(1)This paper proposes a new energy-balanced topology method(EBTM)for wireless sensor networks(WSNs),based on the modern complex network theory.Due to the limited energy and transmission ability of sensor nodes,the next link node in EBTM is chosen according to not only conventional features—the residual energy and the node degree,but also new features—the transmission energy consumption and the transmission pressure between two nodes.Hence,a flow pressure estimation model is established and inconsistent initial edge weights are introduced in this paper,which result in various nodes-selected strategies in different regions.Moreover,in order to better adapt to diverse demands in reality,tunable parameters for the new method is proposed.In the experiments,EBTM is compared with BA,EAEM and FASF,and the experimental results show that EBTM outperforms the other algorithms with longer network lifetime and better robustness.(2)This paper proposes a new link-adding strategy for WSNs called local link-adding strategy(LLA).Different from the previous strategies,in LLA strategy,local theory and characters of WSNs are considered.Thus,different sets are distinguished by tunable parameters.In order to escape some extreme cases in IE strategy,twice sorting is used in LLA.Simulations reveal that LLA has better robustness with LBF and IE,also stays a high traffic capacity,which are more necessary requirements for large scale WSNs.